Re: [Qemu-devel] [PATCH v2 00/11] Point-in-time snapshot exporting over NBD

2013-07-22 Thread Paolo Bonzini 
Il 23/07/2013 03:52, Wenchao Xia ha scritto:
> >Dirty change tracking is what miss for a full solution, which
> > naturally exist in backing chain snapshot.
> 
>   Any one is doing dirty tracking feature? If no, I'd like to draft a
> version based on add-cow.

I had started it, I can dig it up.  Remind me if I don't write about it
in a couple of days.

Paolo

Re: [Qemu-devel] [PATCH v2 00/11] Point-in-time snapshot exporting over NBD

2013-07-22 Thread Wenchao Xia 
于 2013-7-19 18:41, Wenchao Xia 写道:
> 于 2013-7-17 17:42, Fam Zheng 写道:
>> This series adds for point-in-time snapshot NBD exporting based on
>> blockdev-backup (variant of drive-backup with existing device as target). 
>> This
>> patch is built on top of imain's sync mode patches for drive-backup.
>>
>> We get a thin point-in-time snapshot by COW mechanism of drive-backup, and
>> export it through built in NBD server. The steps are as below:
>>
>Compared with driver-backup, this approach is more flexible in some way,
> since user can read the content and copy out what he need, with a
> sacrifice of write of old data to temporary file. Overall it seems good.
> 
>Dirty change tracking is what miss for a full solution, which
> naturally exist in backing chain snapshot.
> 
>>
> 
> 
  Any one is doing dirty tracking feature? If no, I'd like to draft a
version based on add-cow.

-- 
Best Regards

Wenchao Xia

Re: [Qemu-devel] [PATCH v2 00/11] Point-in-time snapshot exporting over NBD

2013-07-19 Thread Wenchao Xia 
于 2013-7-17 17:42, Fam Zheng 写道:
> This series adds for point-in-time snapshot NBD exporting based on
> blockdev-backup (variant of drive-backup with existing device as target). This
> patch is built on top of imain's sync mode patches for drive-backup.
> 
> We get a thin point-in-time snapshot by COW mechanism of drive-backup, and
> export it through built in NBD server. The steps are as below:
> 
  Compared with driver-backup, this approach is more flexible in some way,
since user can read the content and copy out what he need, with a
sacrifice of write of old data to temporary file. Overall it seems good.

  Dirty change tracking is what miss for a full solution, which
naturally exist in backing chain snapshot.

> 


-- 
Best Regards

Wenchao Xia

[Qemu-devel] [PATCH v2 00/11] Point-in-time snapshot exporting over NBD

2013-07-17 Thread Fam Zheng 
This series adds for point-in-time snapshot NBD exporting based on
blockdev-backup (variant of drive-backup with existing device as target). This
patch is built on top of imain's sync mode patches for drive-backup.

We get a thin point-in-time snapshot by COW mechanism of drive-backup, and
export it through built in NBD server. The steps are as below:

 1. (SHELL) qemu-img create -f qcow2 BACKUP.qcow2 

(Alternatively we can use -o backing_file=RUNNING-VM.img to omit explicitly
providing the size by ourselves, but it's risky because RUNNING-VM.qcow2 is
used r/w by guest. Whether or not setting backing file in the image file
doesn't matter, as we are going to override the backing hd in the next
step)

 2. (HMP) drive_add backing=ide0-hd0,file=BACKUP.qcow2,id=target0,if=none

(where ide0-hd0 is the running BlockDriverState name for
RUNNING-VM.img. This patch implements "backing=" option to override
backing_hd for added drive)

 3. (QMP) blockdev-backup device=ide0-hd0 sync=none target=target0

(this is the QMP command introduced by this series, which use a named
device as target of drive-backup)

 4. (QMP) nbd-server-add device=target0

When image fleecing done:

 1. (QMP) block-job-complete device=ide0-hd0

 2. (HMP) drive_del target0

 3. (SHELL) rm BACKUP.qcow2

v2:
* Introduce soft and hard reference count, compared to a single type of ref 
count in v1.
* Add "backing=" option to drive_add.
* Introduce QMP command "blockdev-backup"

Fam Zheng (11):
  block: replace in_use with refcnt_soft and refcnt_hard
  block: use refcnt for bs->backing_hd and bs->file
  block: use refcnt for drive_init/drive_uninit
  block: use refcnt for device attach/detach
  migration: omit drive ref as we have bdrv_ref now
  xen_disk: simplify blk_disconnect with refcnt
  block: hold hard reference for backup/mirror target
  block: simplify bdrv_drop_intermediate
  block: add assertion to check refcount before deleting
  block: add option 'backing' to -drive options
  qmp: add command 'blockdev-backup'

 block-migration.c   |   6 +-
 block.c | 155 +---
 block/backup.c  |   3 +-
 block/blkdebug.c|   1 +
 block/blkverify.c   |   1 +
 block/mirror.c  |   4 +-
 block/snapshot.c|   2 +-
 block/stream.c  |   2 +-
 block/vvfat.c   |   1 +
 blockdev.c  |  77 +++-
 blockjob.c  |   6 +-
 hw/block/dataplane/virtio-blk.c |   4 +-
 hw/block/xen_disk.c |   7 +-
 include/block/block.h   |   5 +-
 include/block/block_int.h   |   3 +-
 qapi-schema.json|  49 +
 qmp-commands.hx |  22 ++
 17 files changed, 248 insertions(+), 100 deletions(-)

-- 
1.8.3.2